1. Field of the Invention
The present invention relates to a color video signal processing device and, more particularly, to a color video signal processing device for performing the encoding on a block unit basis of (n.times.m) samples (n and m being integers of 2 or more).
2. Related Background Art
In general, in a color video signal consisting of a luminance signal and two kinds of color difference signals, the band of the color difference signals is narrower than the band of the luminance signal and the sampling frequency upon digitization is also set to about 1/4 of the sampling frequency of the luminance signal.
Further, when considering the visual characteristic on a screen, even when the information amount of a chrominance signal is further compressed as compared with the information amount of the luminance signal, the deterioration in picture quality is inconspicuous. However, even if the sampling frequency of the color difference signal is simply further reduced, the deterioration in resolution in the horizontal direction becomes conspicuous.
Therefore, hitherto, there has been proposed a method whereby the information amount is reduced by half by line sequencing two kinds of color difference signals (C.sub.N, C.sub.W) or a method whereby the information amount is reduced by half by an offset subsampling such that pixels shifted between lines or fields are transmitted and thinned out to the other pixels.
On the other hand, in recent years, the video signal has further become finer and finer and a test of what is called a HD (High Definition) television signal having 1000 or more scanning lines has also been executed. Consequently, there is a tendency for the information amount of the video signal further to increase. In the case of considering the limitation of the transmitting rate of a transmission path, the information amount must be further compressed. For this purpose, various kinds of highly efficient encoding systems have been proposed.
As an example of the highly efficient encoding systems, there has been known a block encoding in which one picture plane is divided into a plurality of encoding blocks each consisting of (n.times.m) sample points and information is compressed by using the correlation among the pixels in each block without causing the picture quality to be deteriorated. According to such a block encoding, since the encoding can be executed by using the pixels having the highest correlation, it is advantageous in terms of the deterioration in picture quality remaining small and the propagation of code errors being suppressed only in each block.